Two phytoplasmas elicit different responses in the insect vector Euscelidius variegatus Kirschbaum

(1)

28 July 2021

AperTO - Archivio Istituzionale Open Access dell'Università di Torino

Original Citation:

Two phytoplasmas elicit different responses in the insect vector Euscelidius variegatus Kirschbaum

Published version:

DOI:10.1128/IAI .00042-18 Terms of use:

Open Access

(Article begins on next page)

Anyone can freely access the full text of works made available as "Open Access". Works made available under a Creative Commons license can be used according to the terms and conditions of said license. Use of all other works requires consent of the right holder (author or publisher) if not exempted from copyright protection by the applicable law. Availability:

This is the author's manuscript

(2)

1

Two phytoplasmas elicit different responses in the insect vector Euscelidius variegatus 1

Kirschbaum 2

3

Running title: Vector-borne plant pathogens modulate insect immunity 4

5

Luciana Galettoa#, Simona Abbàa, Marika Rossia, Marta Vallinoa, Massimo Pesandoa, Nathalie 6

Arricau-Bouveryb, Marie-Pierre Dubranab, Walter Chitarraa,c, Mattia Pegoraroa, Domenico Boscoa,d, 7

Cristina Marzachìa 8

9 a

Istituto per la Protezione Sostenibile delle Piante, CNR, National Research Council of Italy, IPSP-10

CNR, Torino, Italy 11

b

INRA, Univ. Bordeaux, UMR Biologie du Fruit et Pathologie, UMR 1332, Villenave d'Ornon 12

Cedex, France 13

c_{Centro di Ricerca per la Viticoltura e l’Enologia, CREA, Council for Agricultural Research and} 14

Economics, CREA-VE, Conegliano (TV), Italy 15

d

Dipartimento di Scienze Agrarie, Forestali ed Alimentari DISAFA, Università degli Studi di 16

Torino, Grugliasco (TO), Italy 17

18

#Corresponding author: Luciana Galetto, luciana.galetto@ipsp.cnr.it 19

20

IAI Accepted Manuscript Posted Online 12 March 2018 Infect. Immun. doi:10.1128/IAI.00042-18

on April 11, 2018 by INRA - Institut National de la Recherche Agronomique

http://iai.asm.org/

(3)

2 ABSTRACT

21

Phytoplasmas are plant pathogenic bacteria transmitted by hemipteran insects. The leafhopper 22

Euscelidius variegatus is a natural vector of chrysanthemum yellows phytoplasma (CYp) and a

23

laboratory vector of Flavescence dorée phytoplasma (FDp). The two phytoplasmas induce different 24

effects on this species: CYp slightly improves, while FDp negatively affects insect fitness. To 25

investigate the molecular bases of these different responses, RNA-seq analysis of E. variegatus 26

infected with either CYp or FDp was performed. The sequencing provided the first de novo 27

transcriptome assembly for a phytoplasma vector, and a starting point for further analyses on 28

differentially regulated genes, mainly related to immune system and energy metabolism. Insect 29

phenoloxidase activity, immunocompetence, and body pigmentation were measured to investigate 30

the immune response, while respiration and movement rates were quantified to confirm the effects 31

on energy metabolism. The activation of insect immune response upon FDp infection, which is not 32

naturally transmitted by E. variegatus, confirmed that this bacterium is mostly perceived as a 33

potential pathogen. Conversely, the acquisition of CYp, which is naturally transmitted by E. 34

variegatus, seems to increase the insect fitness by inducing a prompt response to stress. This

long-35

term relationship is likely to improve survival and dispersal of the infected insect, thus enhancing 36

the opportunity of phytoplasma transmission. 37

38

INTRODUCTION 39

Phytoplasmas are wall-less plant pathogenic bacteria of the class Mollicutes, that cause yield losses 40

in many crops worldwide. They colonize plant phloem tissues and are transmitted by phloem-41

feeding hemipterans (1). In insects, ingested phytoplasmas cross the gut, multiply in the haemocoel 42

and invade salivary glands, before being transmitted during feeding on a new plant (2). Currently, 43

genomes of four phytoplasmas are fully sequenced and annotated and few others are available as 44

drafts (1). Phytoplasmas have small genomes (~750 kb) due to a genome reduction that resulted in 45

the loss of important metabolic pathways: as a consequence, these intracellular bacteria depend on 46

http://iai.asm.org/

(4)

3

their hosts for many essential metabolites (3). Due to these strict interactions with hosts and to the 47

difficulty of their axenic cultivation, phytoplasmas need to be studied directly in their hosts. Little is 48

known about pathogenicity mechanism, even if some pathogen-secreted virulence factors have been 49

identified, mainly in strains of ‘Candidatus Phytoplasma asteris’ (2). Studies on this species suggest 50

that phytoplasmas are able to modulate their gene expression during host switching between plant 51

and insect (4, 5). 52

Flavescence dorée (FD) is an important grapevine disease caused by a 16SrV phytoplasma, mainly 53

transmitted, under field conditions, by the hemipteran cicadellid Scaphoideus titanus. FD 54

phytoplasma (FDp) is a plant quarantine pathogen in the European Union and represents one of the 55

major threats to southern European viticulture. Vitis vinifera and S. titanus do not represent ideal 56

experimental organisms for laboratory tests: grapevine is a perennial woody plant and the 57

leafhopper is a monovoltine species. Thus, a laboratory model has been established to manage FDp 58

infection cycle with the herbaceous Vicia faba as plant and the polivoltine leafhopper Euscelidius 59

variegatus as vector (6).

60

Chrysanthemum yellows phytoplasma (CYp), 16SrI-B ‘Ca. P. asteris’, is associated with a disease 61

of ornamental plants in north-western Italy, where E. variegatus is one of the most important 62

natural vectors (7). Like FDp, CYp infections can be obtained under controlled conditions with 63

Chrysanthemum carinatum as host plant and E. variegatus as vector. The two phytoplasmas have

64

opposite effects on the vector fitness: FDp significantly reduces insect longevity and fecundity (8), 65

meanwhile CYp induces a slight fitness increase (9). CYp shows greater ability than FDp to 66

colonise the salivary glands of the vector and therefore it is more efficiently transmitted by E. 67

variegatus (7). So far, physical maps of FDp and CYp genomes, drafts of their genome sequences

68

and an FDp transcriptome analysis (10) are available. By contrast, neither the genome nor the 69

transcriptome of E. variegatus is available. Few proteins of the insect have been identified, such as 70

in vitro interacting partners of CYp antigenic membrane protein (Amp) (11), which is necessary for

71

CYp acquisition by insect vectors (12). The specificity of FDp transmission is presumably mediated 72

http://iai.asm.org/

(5)

4

by variable membrane protein A (VmpA), a phytoplasma protein that is supposed to interact with 73

insect tissues and shows high sequence variability in different strains transmitted by different vector 74

species (13). 75

In a few studies the plant response to phytoplasma infection was addressed with NGS 76

transcriptomic approaches (1), but to our knowledge this technique has never been used to 77

investigate either the transcriptomic profile of an insect vector infected by these pathogens or the 78

alterations induced by different phytoplasmas in the same host. Here we investigated the effects of 79

two genetically different phytoplasmas on the same insect vector, providing: i) de novo assembly of 80

E. variegatus transcriptome, ii) differential expression analysis of E. variegatus transcripts under

81

infection with CY or FD phytoplasmas iii) validation of the differential expression profiles and iv) 82

biological experiments to support the transcript profiling results. 83

84

RESULTS 85

RNAseq and differential gene expression. Diagnostic RT-qPCR assays confirmed the presence of 86

CY and FD phytoplasmas in E. variegatus samples (Eva_CY and Eva_FD). The phytoplasma 87

populations, expressed as mean phytoplasma 16S/insect18S ratio, were 4.6703 (SEM ± 9.84 E-88

04) and 2.08 E-04 (SEM ± 6.54 E-05) for insects infected with CYp and FDp, respectively. 89

Analyses of the cDNA libraries obtained from Eva_CY and Eva_FD resulted in a combined de 90

novo assembly comprising around 135,000 transcripts, with an average GC content of 40%, a

91

median contig length of 433 bp, and an average contig length of 833 bp. Due to the lack of genomic 92

sequence information of E. variegatus, the functional annotation of transcripts was conducted using 93

blastx against the NCBI non-redundant (nr) database. The Blast2GO platform was then used to 94

assign the Gene Ontology (GO) terms to the predicted proteins with known function. The results of 95

the de novo assembly and the following transcript annotations are summarized in Table S1. The 96

species distributions of the best blastx matches for each sequence are shown in Table S2. 97

http://iai.asm.org/

(6)

5

Transcriptome profiles of E. variegatus infected by the beneficial CY or the pathogenic FD 98

phytoplasmas were compared to elucidate the differential vector responses. 99

Differential expression analysis revealed that 84 transcripts were up-regulated and 13 down-100

regulated in Eva_CY in comparison with Eva_FD (Tables 1 and 2, S3 and S4). The up-regulated 101

genes could be classified in few main functional categories: immune response (11 transcripts), 102

movement and energy metabolism (34 transcripts), proteases (9 transcripts), extracellular matrix (20 103

transcripts), nucleic acid binding (6 transcripts), and detoxification (4 transcripts). Down-regulated 104

genes could be ascribed to immune response (10 transcripts), movement and energy metabolism (1 105

transcript), proteases (1 transcript), and detoxification (1 transcript) functional categories. Some of 106

these putative metabolic functions were further investigated to explore the phenotypes correlated 107

with the altered gene regulation. 108

Cellular and humoral immunity in response to phytoplasma infections. To investigate the 109

effects of the phytoplasma presence on immune response, gene expression analysis of selected 110

transcripts, enzymatic activity and biological assays were performed (Fig. 1A; Tables S5, S6, S7). 111

Healthy controls (Eva_H), not exposed to phytoplasmas and PCR negative, were included in the 112

following experiments to highlight the mechanisms underpinning the differential effects on E. 113

variegatus immune response upon infection with the two phytoplasma species. Gender was taken

114

into account, but whenever no sex-related differences were recorded within the same category data 115

were pooled. RT-qPCR validation was run on 42 samples (each made up of five pooled insects), 116

phenoloxidase activity was measured for 36 samples (each made up of haemolymph collected from 117

five insects), pigmentation and immunocompetence assays were tested on 170 and 46 specimens, 118

respectively. 119

Gene expression. Kazal-type 1 serine protease inhibitor and phenoloxidase genes were selected

120

from the RNAseq results and literature search (14), respectively, and analysed by RT-qPCR in 121

CYp-, FDp-infected and healthy insects (Eva_H, _CY, _FD). Similar levels of phenoloxidase 122

transcripts were recorded in the insects, regardless of the sex and the infection status (Fig. 2A), 123

http://iai.asm.org/

(7)

6

whereas Kazal-type 1 transcripts were significantly more abundant in Eva_FD males compared to 124

Eva_CY ones (P=0.022) (Fig. 2B). Transcripts of the same gene were significantly more abundant 125

in healthy females compared to healthy (P=0.002) and CYp-infected males (P=0.023) (Table S5). 126

Up-regulation of Kazal-type 1 serine protease inhibitor in Eva_FD confirmed the differential 127

expression obtained by RNAseq analysis (Table 2). 128

Enzymatic activity. Phenoloxidase (PO) and Pro-PO activities associated with haemocytes and the

129

plasma fraction of haemolymph were quantified. The specific inhibitor phenylthiourea inhibited PO 130

and ProPO activities in all assays with the exception of PO in haemocytes (data not further 131

analyzed). The PO activities of the plasma fractions were similar irrespective of sex and infection 132

status (Table S6). The plasma ProPO activities were significantly lower in males than in females, 133

irrespective of the infection status (P<0.001, P<0.001 and P=0.018, for H, CY and FD, 134

respectively); this activity was higher in FDp- compared to CYp-infected insects (two-fold 135

increase), although the difference was significant only for males (P=0.046) (Fig. 2C). ProPO 136

activities were similar in haemocytes from females and males of each infection status, whereas the 137

enzymatic activity of Eva_FD was double than that of Eva_CY (P=0.017) (Fig. 2D). Steeper slopes 138

of the linear phases of Eva_FD assays further confirmed faster enzymatic reactions in FDp-infected 139

insects (Table S6). 140

Pigmentation assay. Since PO activity may be correlated to cuticular colour (15), pigmentation of

141

bodies (dorsal side) and forewings of healthy and phytoplasma-infected E. variegatus was further 142

explored. Pigmentation was expressed as grey intensity: 0 for black to 255 for white (Fig. 2E). 143

Considering the three experimental conditions (Eva_H, _CY, _FD), males always showed 144

significantly darker bodies and wings than females (P=0.005 for healthy bodies and P<0.001 for all 145

other comparison) (Table S7). Wing pigmentation did not show any significant variation in the 146

presence of phytoplasmas, while bodies of FDp-infected female and male insects were significantly 147

darker than healthy and CYp-infected ones (female: P<0.001 for both comparison; male: P<0.001 148

and P=0.003 for H vs. FD and CY vs. FD, respectively) (Fig. 2E). 149

http://iai.asm.org/

(8)

7

Immunocompetence assay. To determine whether the phytoplasma presence could influence insect

150

immune response, nylon threads were implanted in insect abdomens to measure melanization as 151

well as the number of haemocytes adhered to the threads (Fig. 2F, Table S7). There was no 152

influence of sex on melanization index (MI) and number of cells adherent to the nylon threads, 153

irrespective of the infection status, but there was a significant difference of MIs of nylon threads 154

implanted in CYp-infected E. variegatus compared to those implanted in healthy insects (Fig. 2F, 155

bars). Up to five times more haemocytes adhered to nylon threads implanted into CYp-infected E. 156

variegatus compared to those implanted into H and FDp-infected insects, and this difference was

157

significant in both cases (P<0.001) (Fig. 2F, dots). 158

Movement and energy metabolism. Genes involved in muscle contraction and synthesis of 159

intermediate metabolites for energy production were selected among those differentially regulated 160

according to the RNAseq results and literature search (16) (Fig. 1B; Table S5). To investigate the 161

effects of phytoplasma infection on insect mobility and respiration rate, several parameters were 162

measured (Fig. 3F, G, H), (Table S8). Healthy controls (Eva_H) were included in the following 163

experiments to better describe the metabolic response of E. variegatus challenged by the two 164

phytoplasmas, and whenever no sex-related differences were recorded within the same category 165

data were pooled. RT-qPCR validation was run on 42 samples (each made up of five pooled 166

insects), movement assay was tested on 138 specimens and CO2 production was measured in 24 167

insect groups (each made up of three specimens). 168

Gene expression. Myosin light chain, tropomyosin, arginine kinase and maltase were analysed by

169

RT-qPCR in insects of the three experimental conditions (Eva_H, _CY, _FD). The four analysed 170

transcripts were significantly more abundant in males than in females, regardless of the infection 171

status, with the exception of tropomyosin and arginine kinase in healthy insects (Fig 3A-D, Table 172

S5). In general, CYp-infected males showed higher transcript levels compared to healthy and FDp-173

infected males. These differences were significant only for tropomyosin, with about three times 174

more transcripts in CYp-infected vs. healthy males (P=0.018) and for arginine kinase, with about 175

http://iai.asm.org/

(9)

8

twice more transcripts in CYp-infected males vs. healthy and FDp-infected ones (P<0.001 for both 176

comparison). Arginine kinase transcripts were also significantly more abundant in CYp-infected 177

females compared with FDp-infected ones (P=0.024) (Fig. 3C). Up-regulation of arginine kinase 178

and maltase in Eva_CY confirmed the differential expression results of the RNAseq analysis (Table 179

1). 180

Protein expression. To further characterize the differential expression of proteins involved in

181

movement, Western blot analysis was performed on healthy and phytoplasma-infected insects with 182

anti-tropomyosin commercial antisera (Figs. 3I and S1). Tropomyosin was more expressed in males 183

than in females, regardless of the infection status, therefore confirming the transcriptional analyses. 184

Nevertheless, there were no evident differences in protein expression levels among Eva_H, _CY 185

and _FD categories. 186

Movement and respiration functional assays. Movement parameters (permanence time in a two

187

circle arena and numbers of jumps) showed neither significant variation between males and 188

females, nor among the three experimental conditions, although a trend of a faster movement was 189

observed for CYp-infected males in comparison with healthy and FDp-infected ones (Fig. 3E, bars; 190

Table S8). Consistently, measurements with the gas analyzer showed that Eva_CY produced a 191

significantly higher amount of CO2 than Eva_H and _FD, irrespective of sex (P=0.002 and P=0.027 192

for H vs. CY and CY vs. FD, respectively) (Fig. 3E, dots; Table S8). 193

Protease regulation. Cathepsin L was selected to investigate the effect of phytoplasma infection on 194

protease regulation. This protein is the major component of the gut digestive enzymes in many 195

invertebrates (17) and, among other functions, is involved in controlling symbiont populations (18). 196

Healthy controls (Eva_H) were included in the following experiments, to dissect the effects of the 197

different phytoplasmas on E. variegatus protease regulation. 198

Gene expression. The expression profile of four isoforms (namely 92i3, 92i4, 92i6 and 473) of

199

cathepsin L was analysed by RT-qPCR in insects of the three experimental conditions (Eva_H, 200

_CY, _FD). Isoform 473 was the most strongly up-regulated in Eva_CY within the protease 201

http://iai.asm.org/

(10)

9

category (Table 1), whereas isoforms 92i3, 92i4, 92i6 were selected since, among all E. variegatus 202

transcripts annotated as “cathepsin L”, they showed the highest identity with the immunogenic 203

peptide recognized by the commercial anti-cathepsin L antibody used for the Western blot. 204

Significant differences between female and male insects were present for isoform 473, irrespective 205

of the infection status (up-regulated in females, P=0.007, P<0.001 and P<0.001, for Eva_H, CY and 206

FD, respectively) and for isoforms 92i3 and 92i6 specifically for FDp-infected E. variegatus (up-207

regulated in males, P=0.043 and P=0.017 for 92i3 and 92i6, respectively) (Fig. 4A-D and Table S5). 208

Transcripts of isoforms 92i3 and 92i4 were significantly up-regulated in CYp-infected females vs. 209

FDp-infected ones (P=0.012 and P=0.023 for 92i3 and 92i4, respectively) (Fig. 4A and B). Those of 210

isoform 473 were twice more abundant in CYp-infected females vs. healthy and FDp-infected ones, 211

and these differences were significant (P=0.003 and P=0.012, for H vs. CY and CY vs. FD, 212

respectively) (Fig. 4D). Up-regulation of cathepsin L_473 in Eva_CY compared to Eva_FD 213

confirmed the results of the RNAseq analysis (Table 1). 214

Putative protein characterization. The predicted amino acid sequences of the four cathepsin L

215

isoforms were analysed (Fig. 4E and F). All four isoforms showed putative signal peptide from aa 216

1-16, a pro-region containing the propeptide inhibitor domain and a predicted mature protein 217

including the cysteine protease domain. Putative glycosilation sites were predicted on different 218

isoforms: 2 nitrogen-linked and 5 oxygen-linked. Isoform 92i6 showed a unique putative N-linked 219

site at position 37 and the isoform 473 missed an O-linked site at position 125. Isoforms 92i4 and 220

92i6 showed highly similar pre-proteins and identical mature forms, whereas isoforms 473 and 92i3 221

were the most diverse ones (Fig. 4F). The immunogenic peptide, recognized by the commercial 222

antibody used for Western blot, was more similar to isoform 473 than the other ones (Fig. 4F). 223

Protein expression. To further characterize the effect of phytoplasma infection on the expression of

224

cathepsin L protein, Western blot was performed on healthy and phytoplasma-infected insects with 225

anti-cathepsin L commercial antisera (Figs. 4G and S1). The antibody was raised to detect both the 226

pre-protein and the mature form, as it reacts against the immunogenic peptide indicated in Fig. 4E. 227

http://iai.asm.org/

(11)

10

Indeed, the Western blots with anti-cathepsin L antibody (Fig. 4G) showed a complex pattern: two 228

faint bands with high MW (around 37-35 kDa) possibly corresponding to pre-proteins, two intense 229

bands with low MW (around 27-25 kDa) possibly corresponding to the mature forms. Similar band 230

patterns were observed from total proteins of E. variegatus females, irrespective of the infection 231

status. A similar profile, although less intense, was also evident for healthy males. Surprisingly, 232

almost no signal from the lowest MW bands (25 kDa) was detected from phytoplasma-infected 233 males. 234 235 DISCUSSION 236

Relationships between phytoplasmas and their vectors may be pathogenic, neutral or mutualistic 237

(9). CYp and FDp establish different types of relationships with their vector E. variegatus: the 238

former slightly improves vector fitness, the latter is pathogenic. The transcriptional landscape of E. 239

variegatus infected with the two different phytoplasma species was analysed focusing on long

240

lasting modifications occurring in insects during the response to chronic phytoplasma infection, 241

thus avoiding any possible differences related to CYp and FDp multiplication dynamics during 242

early stages of infection. Sex-specific effects were recorded for several of the tested parameters, as 243

already described for immune response (19, 20) as well as insect movement and dispersal (21). 244

Among insects of the Cicadellidae family, few de novo transcriptome assemblies are available: 245

some were obtained from specific insect tissues, namely salivary glands of Nephotettix cincticeps 246

(22) and Empoasca fabae (23) and intestinal tract of Empoasca vitis (24), others from whole bodies, 247

such as those of Graminella nigrifrons (25), Homalodisca vitripennis (26) and Zyginidia pullula 248

(27). Interestingly, the transcriptomic response of G. nigrifrons vector to different plant viruses 249

infections reveals that the expression of cytoskeleton and immunity genes increase in the presence 250

of the persistent propagative rhabdovirus Maize fine streak virus (28). 251

http://iai.asm.org/

(12)

11

Phytoplasma infection modulates insect immune response. Molecular and biological analyses 252

indicate that a different modulation of E. variegatus immune response occurred following FDp 253

infection compared to CYp infection. 254

The altered regulation of the immune system was revealed by RNAseq analysis during infection 255

with both phytoplasmas. Among these, transcripts of the Kazal type 1 serine protease inhibitor were 256

more abundant in FDp-infected insects compared to Eva_CY. Similar serine protease inhibitors 257

have antibacterial activity against bacteria (29) as well as antifungal activity against both plant-258

pathogenic and entomopathogenic fungi, as inhibitors of microbial serine proteases (30). Serpin was 259

the most strongly up-regulated transcript in CYp-infected insects, whereas the snake-like serine 260

protease was among the most strongly down-regulated ones. Clip serine proteases, such as the 261

snake-like ones, are involved in the activation of the ProPO proteolytic cascade in invertebrate 262

immune systems (31, 32), while serpins from different arthropod species inhibit clip domain serine 263

proteases by blocking the activation of ProPO melanization pathway (33, 34). The 264

prophenoloxidase (ProPO) cascade is involved in melanization and encapsulation processes and 265

provides arthropod immunity against bacteria, fungi, protozoan and parasites (34). The opposite 266

regulation of these two transcripts correlates with the lower prophenoloxidase activity and with the 267

less intense cuticular pigmentation observed in CYp-infected compared to FDp-infected insects. 268

Cuticular color is related to immune response in insects (15), and the darker body pigmentation of 269

FDp-infected E. variegatus suggests stimulation of melanization pathway due to a stronger 270

activation of the immune response. The presence of FDp is perceived by the insect as a stress status 271

and therefore it elicits an intense production of melanin. Indeed, prophenoloxidase activities of both 272

plasma and haemocyte lysates were more intense in Eva_FD compared to _CY and _H. On the 273

other hand, the infection with the two phytoplasmas had neither effect on naturally activated 274

phenoloxidase (PO) activity, a good estimate of invertebrate immunocompetence (35), nor on the 275

abundance of the corresponding transcripts. This could be due to the fact that the analyses were 276

performed at late, chronic stages of phytoplasma infection, when colonization of the insect body 277

http://iai.asm.org/

(13)

12

was complete (12). A burst of activated phenoloxidase is, in fact, expected at the onset of the 278

infectious event, as a defence reaction to the immunological challenge (35), as reported for 279

Micrococcus luteus infection of the leafhopper Circulifer haematoceps (36). Surprisingly, when

280

insects are challenged by an additional stress (wounding through nylon thread), the scenario 281

changes. In the immunocompetence assay, insertion of a nylon thread in the insect body mimics a 282

parasite invader and induces encapsulation. The response of FDp-infected insects is similar to that 283

of the healthy ones. On the contrary, CYp-infected insects showed higher MI and higher number of 284

haemocytes, indicating a better capacity of these insects to react to and isolate an invader. E. 285

variegatus is a natural vector of CYp and they share the same ecological niches: these factors could

286

have shaped the insect immune system to fight more promptly against incoming pathogenic 287

organisms. 288

The Kruppel-like factor, a zinc finger DNA-binding protein, is crucial to mediate white spot 289

syndrome virus (WSSV) infection in two different shrimp species (37), and two transcripts of this 290

gene were oppositely regulated in E. variegatus (more abundant in FDp-infected category), 291

suggesting a role for this protein in response to phytoplasma infection. On the other hand, 292

hexamerin transcripts were up-regulated upon CYp infection. Members of this protein family are 293

inducible effector proteins in insect immunity upon bacteria ingestion and have a putative role in 294

gut repair (38). Moreover, in the closely related mollicute-leafhopper association (C. 295

haematoceps/Spiroplasma citri) hexamerin is up-regulated following infection and is required for

296

vector survival after spiroplasma inoculation (36). Besides their role in energy metabolism (see 297

below), arthropod arginine kinases (AK) are also involved in stress response and innate immunity: 298

Apis cerana AK is induced by abiotic and biotic stresses (39), pacific oyster AK modulates

299

bactericidal immune response in haemolymph (40), and AK from shrimp Fenneropenaeus chinensis 300

has been hypothesized as putative receptor of the WSSV virus envelope protein (41). AK transcript 301

was up-regulated upon CYp infection, suggesting different potential roles for this protein during 302

infection with the two phytoplasmas. 303

http://iai.asm.org/

(14)

13

Disulfide bonds are redox-controlled switches for pathogens invasion, and are involved in 304

regulating pathogen entry into the endocytic pathway of vertebrate (42) and some invertebrates (43, 305

44). Recently, a vesicle-mediated colonization of salivary glands has been suggested for CYp 306

infection of E. variegatus (12). Consistently, the protein disulfide-isomerase (five transcripts) and 307

the gamma-interferon inducible lysosomal thiol genes were up-regulated upon CYp infection, 308

supporting the involvement of the endocytic pathway in phytoplasma colonization of the host, as 309

described for Leishmania, Listeria and Chlamydia spp (42). Other bacterial pathogens have 310

developed strategies to interfere with host lipidation mechanisms (45). For example, Salmonella 311

enterica and Legionella pneumophyla exploit host prenylation to direct effector proteins to the

312

pathogen containing vacuole of the host cell (46). Interestingly, transcripts of the farnesyl – 313

geranylgeranyl transferases, the key enzyme of the prenylation pathway, were oppositely regulated 314

in E. variegatus (down-regulated in CYp-infected insects) suggesting different alteration of 315

vesicular trafficking upon infection with CYp and FDp. Phytoplasma may modulate the host 316

metabolism through active secretion of effector molecules and the diversity of the effector arsenals 317

among phytoplasmas (2, 47) may explain the opposite transcription profiles of this gene. 318

Chrysanthemum yellows phytoplasma infection increases energy metabolism. Molecular and 319

biological analyses indicate an activation of E. variegatus energy production metabolism and an 320

increased locomotion activity upon CYp infection. According with RNAseq results, movement and 321

energy production metabolism was the functional category with the highest number of gene 322

transcripts altered upon phytoplasma infection. 323

Titin, twitchin and protein unc-89 are members of the giant cytoskeletal kinase family that mediate 324

sensing and transduction of mechanical signals in the myofibril. These big proteins display elastic 325

conformational deformation and regulate muscle tissue in adaptation to external stimuli (48). 326

Several isoforms of these gene transcripts were up-regulated in CYp-infected E. variegatus. These 327

kinases participate in regulating protein turnover in muscle, and, in particular, unc-89 regulates 328

http://iai.asm.org/

(15)

14

ubiquitin-mediated protein degradation, through recruitment of E3 ubiquitin ligases (48), which 329

indeed was also up-regulated upon CYp infection. 330

Two isoforms of PDZ and LIM domain protein 3 were up-regulated in CYp-infected insects. 331

PDZ/LIM genes encode a large group of proteins that play important and diverse biological roles, 332

but that functionally can all influence or be associated with the actin cytoskeleton (49). Ryanodine 333

receptor (RYR) is the main calcium release complex of the sarcoplasmic reticulum involved in the 334

excitation-contraction coupling of muscle cells (50), and the dihydropyridine receptor (DHPR) is 335

the plasma membrane L-type calcium channel involved in opening of the RYR by a calcium-336

induced calcium release mechanism (51). Transcripts of these genes were inversely regulated (RYR 337

up-regulated and DHPR down-regulated) upon CYp infection, suggesting an altered Ca2+ regulation 338

in the cytosol of muscle cells in response to phytoplasmas. Indeed, transcripts of the calcium-339

transporting ATPase sarcoplasmic/endoplasmic reticulum (SERCA) and of the sarcalumenin were 340

up-regulated in CYp-infected insects. The former is a pump involved in translocation of cytosolic 341

calcium into the sarcoplasmic reticulum to allow relaxation of muscle fibers (16), the latter is a 342

calcium-binding protein involved in fine regulation of cellular calcium storage (52). Moreover, 343

transcripts of the main proteins involved in contraction and cytoskeletal motion, tropomyosin, 344

troponin, myosin, actin and dynein (16) were all up-regulated upon CYp infection. The analysis of 345

tropomyosin confirmed the stronger expression in males than in females, as revealed by RT-qPCR 346

and Western blot, but no evident differences were detected among different infection categories (H, 347

CY, FD), possibly due to several isoforms derived from alternative splicing, a well-known 348

phenomenon for this gene (53). Increased insect movement has been observed in some pathogen-349

vector associations, such as Diaphorina citri infected with ‘Candidatus Liberibacter asiaticus’ (54) 350

and Bombyx mori with BmNPV (55). We tested the intriguing hypothesis that CY phytoplasma, 351

which is naturally transmitted by the insect host, can manipulate the vector movement to increase its 352

transmission, but the parameters recorded during the movement assays did not clearly support this 353

hypothesis. Despite that, muscle contraction is also involved in active insect respiration, so the 354

http://iai.asm.org/

(16)

15

higher expression of the above-mentioned genes in CYp-infected insects could be related with an 355

increase of the respiration rate. This was indeed the case, as higher CO2 levels were produced by 356

CYp-infected E. variegatus in the respiration assay compared to FDp-infected and healthy insects. 357

Additionally, the up-regulation of maltase, hydroxybutyrate dehydrogenase and arginine kinase 358

(AK) transcripts in CYp-infected E. variegatus indicates a stronger activation of the energy 359

production metabolism. Altogether these data point to an augmented movement in CYp-infected 360

insect, which may positively influence phytoplasma transmission. 361

Phytoplasma infection alters protease regulation. Cathepsins are proteases generally stored in 362

lysosomes, involved in several processes like development, apoptosis and immunity of arthropods 363

(17, 56). Upon CYp infection, transcripts of cathepsin L were up-regulated and those of cathepsin D 364

down-regulated, suggesting different roles of these enzymes in response to phytoplasma infection. 365

To confirm RNAseq analysis, four isoforms were chosen among all the E. variegatus “cathepsin L” 366

transcripts: 473, derived from differential expression analysis, and 92i3, 92i4, 92i6, showing the 367

highest identity with the immunogenic peptide recognized by the anti-cathepsin L antibody used in 368

Western blot. The up-regulation of cathepsin L was confirmed by RT-qPCR for three of the four 369

isoforms upon CYp infection in comparison with FDp. The same up-regulation was not present at 370

protein level. Molecular weights of cathepsin L mature proteins differed from the theoretical ones 371

and this may be explained by different glycosylation, one of the post translational-processes 372

occurring during cathepsin maturation (57). Indeed the anti-cathepsin L antibody detected proteins 373

of different sizes, and intriguingly the lowest ones were poorly present in healthy males and nearly 374

absent in phytoplasma-infected males. As transcripts of isoform 473 were significant up-regulated 375

in females, these lowest protein bands are presumably its mature forms. Indeed isoform 473 showed 376

one glycosilation site less than other isoforms and could migrate faster. The absence of mature form 377

of this isoform in infected males might indicate that phytoplasma presence could prevent cathepsin 378

L maturation. Expression of this gene is altered upon microbial infection, either in pathogenic 379

combinations, such as Serratia marcescens in the pea aphid Acyrthosiphon pisum (18), and in 380

http://iai.asm.org/

(17)

16

symbiotic associations, such as Burkholderia symbionts ingested by the bean bug Riportus pedestris 381

(58). Indeed, both in pathogenic and mutualistic associations, bacteria need to avoid lysosomal 382

degradation to establish an intracellular association and this could be true also for CY and FD 383

phytoplasmas. 384

385

In conclusion, transcriptomic and phenotypic results shed some light on the molecular mechanisms 386

underlying the different effects of the two phytoplasmas on the insect vector E. variegatus. Our data 387

show that E. variegatus perceives FD as a pathogen, since it activates an immune response. Lack of 388

natural interactions between FD phytoplasma, mainly restricted to Vitis spp., and the laboratory 389

vector E. variegatus, which does not feed on grapevine, may explain perception of this phytoplasma 390

as non-self. On the other hand, the long-lasting interactions between CY phytoplasma and E. 391

variegatus (that are sympatric) might have driven towards a mutualistic relationship.

392

The prompt and aggressive response to the menace of an external pathogen, mimicked by the nylon 393

thread, may be due to an immune priming activated by CYp and together with the increased energy 394

metabolism are likely to provide an ecological advantage to both the vector and the phytoplasma. 395

396

MATERIALS AND METHODS 397

Insect and phytoplasmas. E. variegatus isolate to-1 was collected in Piedmont (Italy) and reared

398

on oat, Avena sativa (L.) (7). Chrysanthemum yellows phytoplasma (CYp) was isolated in Italy, and 399

maintained by insect transmission on daisy, Chrysanthemum carinatum Schousboe (7). Flavescence 400

dorée phytoplasma (FDp) was isolated in Italy, and maintained by insect transmission on broad 401

bean, Vicia faba L. Daisies, broad beans and oats were all grown from seed in greenhouses (59). 402

For each acquisition access period (AAP), the sanitary status of source plants was confirmed by 403

symptom observation and PCR as detailed in (4). 404

Three experimental conditions were set up. Fifth instar healthy nymphs were separately fed on i) 405

healthy daisies and broad beans (Eva_H), ii) CYp-infected daisies (Eva_CY) or iii) FDp-infected 406

http://iai.asm.org/

(18)

17

broad beans (Eva_FD) for an AAP of 7 days and then transferred on oat for a 28-day latency period 407

(LP). At 35 days post acquisition survived insects were sexed, analyzed for phytoplasma infection 408

and used. 409

RNA extraction. Total RNA was extracted from 64 samples (each made of 5 insects), nearly 20 for 410

each experimental condition (10 female and 10 male) using Direct-zol RNA Mini Prep Kit (Zymo 411

Research). RNA was analyzed in a Nanodrop spectophotometer and in a Bioanalyzer 2100 Expert 412

Agilent Technologies, to evaluate concentration, purity and quality of the samples. 413

Phytoplasma detection and quantification. Total RNA was treated with Turbo RNase-free DNase 414

I (Applied Biosystems). For CYp and FDp diagnosis, cDNA was synthesized from total RNA (800 415

ng) using High Capacity cDNA reverse transcription kit (Applied Biosystems). Two l of cDNA 416

were used as template in qPCR with iTaq Universal Probes Supermix (Bio-Rad) and primers 417

CYS2Fw/Rv and TaqMan CYS2Probe (4). The same primers and probe, targeting phytoplasma 418

16SrRNA, and primers MqFw/Rv with TaqMan MqProbe, targeting insect 18SrRNA (4) were used 419

to quantify phytoplasma load. Four serial 100-fold dilutions of p-GemTEasy (Promega) plasmids, 420

harboring portions of ribosomal genes from phytoplasma and insect, were included to calculate 421

phytoplasma16S/insect18S ratio. 422

RNA-seq, differential gene expression and sequence analysis. Six micrograms of RNA extracted 423

from insects fed on phytoplasma-infected plants (Eva_CY; Eva_FD) and showing similar 424

phytoplasma amount were sent to Macrogen (South Korea) for cDNA libraries construction and 425

sequencing, as detailed in (59). Each library was obtained from a pooled sample of 20 males and 20 426

females. To generate a comprehensive landscape of the E. variegatus transcriptome, the datasets 427

generated by the cDNA libraries (two biological replicates for each condition) were pooled, 428

trimmed by Trimmomatic v0.32 (60), quality checked by FastQC v0.9(61), de novo assembled 429

using Trinity v2.0.6 (62) and clustered by cd-hit-est (63) with a sequence identity cut-off = 0.98. 430

Each transcript was analyzed by blastx against the NCBI nr database with a cut-off Expected value 431

of 1e-04. Only transcripts with arthropods as best top hits species were retained for further analysis. 432

http://iai.asm.org/

(19)

18

Functional annotation for each of the selected transcripts was obtained by loading the corresponding 433

XML output files in Blast2GO and running the mapping and the annotation options with default 434

parameters to retrieve GO terms and assign reliable functions, respectively. In addition, sequences 435

were analyzed for orthology predictions with eggnog (64) with DIAMOND mapping mode. Open 436

reading frames (ORF) were predicted by TransDecoder (65) using single_best_orf” and “--437

retain_pfam_hits” options, which allow to retain only the single best ORF for each transcript 438

according to the presence of a significant Pfam hit. 439

Reads were loaded to NCBI's Sequence Read Archive (SRA) database with the following accession 440

numbers: SRR5816888, SRR5816889, SRR5816890, SRR5816891. 441

For differentially expressed gene (DEG) identification, DESeq2 package (66) v. 1.14.1 was run on a 442

60 core and 256 GB RAM local machine, running Ubuntu server 12.04 LTS. DEG selection was 443

based on an adjusted p-value ≤ 0.01 and a log2FC (Fold Change) ≥ 0.5 for up-regulated genes and ≤ 444

-0.5 for down-regulated genes. 445

SignalP 4.1 (67), Prosite (68) and GlycoEP (69) were used to predict putative signal peptide, active 446

and glycosylation sites on cathepsin L isoforms, respectively. KEGG pathway database was used 447

for Fig. 1B preparation (70). 448

qPCR validation.

449

Some genes were selected from the RNAseq results and literature search and analyzed by RT-qPCR 450

in CYp-, FDp-infected and healthy insects (Eva_H, _CY, _FD). Reverse transcriptase reactions 451

were performed on the RNA extracted from 42 samples (each made up of five pooled insects): 452

seven samples of males and seven of females for each of the three conditions. These samples 453

included those used for library construction as well as new ones. Complementary DNA was used as 454

template for qPCR with primers in Table S9 and iTaq Universal SYBRGreen Supermix (Bio-Rad) 455

with an annealing/extension temperature of 60°C. Primers were checked to target unique isoform in 456

the whole E. variegatus transcriptome. Among the six putative reference genes tested, the insect 457

elongation factor-1α, glutathione S-transferase and heat shock protein 70-1 were selected as the 458

http://iai.asm.org/

(20)

19

most stable under the three conditions (Eva_H, Eva_FD and Eva_CY) (Table S10) and used for 459

qPCR gene expression analysis, according to (71). Normalized relative quantities for each condition 460

were compared. 461

Phenoloxidase activity. The enzymatic activity of naturally activated phenoloxidase (PO) and 462

proenzyme prophenoloxidase (ProPO) were measured in plasma and haemocyte lysate supernatant 463

(HLS) as described (35, 36). About 5 haemolymph samples were tested for each sex and condition. 464

The optical density (OD) at 490 nm was determined immediately, after 30 min and then every hour 465

for 15 h using a Bio-Rad Microplate Spectrophotometer. One unit of activity was defined as a 466

change of 0.001 OD490nm per minute in the linear phase of reaction. Specificity was tested using 467

phenylthiourea (Sigma, 4 mg/ml) to inhibit enzyme activity. 468

Pigmentation assay. The pigmentation of forewing and body (dorsal side) devoid of appendices 469

was calculated through image analysis for about 50 insects for each condition. Images were taken 470

under a stereomicroscope with a D5000 Nikon controlled by Camera Control Pro 2 software and 471

analysed with Fiji software (72). The outline of the object to be measured was marked by the 472

freehand selection tool. Light conditions, camera and software setting were not changed throughout 473

image acquisition of the whole set of samples. Nevertheless, measure of each object was normalized 474

against a white area used as internal standard. Mean degree of grey intensity was expressed as a 475

numerical reading ranging from 0 for black to 255 for white. 476

Immunocompetence assay. A nylon thread (length 2-4 mm, Ø 80 m) was implanted in abdomen

477

of CO2 anaesthetized insects under a stereoscope. About 50 insects were treated for each condition. 478

Insects were transferred to oat for 72 h, collected and dissected to recover the nylon thread in 900 µl 479

10% PBS. Following overnight fixation at 4°C (4% paraformaldehyde, 0.1% Triton X100 in 10% 480

PBS), the threads were washed, DAPI stained and photographed under light and UV microscope. 481

Three images were taken at different z axes, to ensure the best count of nuclei of cell adherent to the 482

thread. Image analyses was performed with Fiji software (72), and Melanization Index (MI) was 483

calculated as the ratio between the integrated density per surface unit of nylon portions inside and 484

http://iai.asm.org/

(21)

20

outside the body of each insect. About 15 insects were analysed for each condition. Number of 485

adherent cells were calculated by summing the DAPI stained nuclei in the three pictures of each 486

thread. 487

Western blots. For each category, proteins were extracted from four samples (each made up of five 488

pooled insects), quantified by Bradford reagent (Bio-Rad), and load on 12% polyacrylamide gels 489

(12 g/lane), together with pre-stained and unstained broad range standards (Bio-Rad) (11). Gels 490

were either stained with colloidal Coomassie or blotted on PVDF membrane. Membranes were 491

blocked for 1 h (3% BSA in TBS 0.1% Tween, BSA-TBST), incubated overnight at 4°C with 492

primary antibodies (ab50567 rat-developed tropomyosin, and ab200738 rabbit-developed anti-493

cathepsin L, Abcam plc) both diluted 1:1000 in BSA-TBST, washed, incubated 2 h with 494

corresponding horseradish peroxidase conjugated secondary antibodies (A4416 GAM-HRP, and 495

A0545 GAR-HRP, Sigma, respectively) both diluted 1:10000 in BSA-TBST, washed and 496

developed with West Pico SuperSignal chemiluminescent substrate (Pierce) in a VersaDoc 4000 497

MP (Bio-Rad). Each experiment was repeated three times. 498

Movement and respiration assays. To evaluate insect movement among the three conditions, 499

insects were anaesthetized for 30 sec and put one at a time in the middle of two concentric circles 500

(Ø 2 and 6 cm) drawn on a paper, covered with a glass cylinder (height 20 cm) and continuously 501

observed for 5 min. Time required to leave the two circles and numbers of jumps were recorded. 502

About 20 insects were tested for each sex and condition. 503

To evaluate insect respiration, CO2 production was monitored within the standard “broad leaf 504

chamber” of a LCpro+ (ADC BioScientific) gas analyzer, as described for Drosophila (73). To 505

measure gas exchange, groups of three adults (same sex and same category) were put in a mini-cage 506

(1.5 ml tube, deprived of bottom and sealed with net). To allow better survival, 200 l of feeding 507

solution (12) were put in the mini-cage cap and covered by a parafilm layer. For each sex and 508

category, 4 groups were analysed. Each mini-cage was left 30 min in the chamber before measure 509

http://iai.asm.org/

(22)

21

(5 reading replicates). The CO2 production, expressed by the analyzer in mol/sec m2, was 510

transformed in l/h per insect according with (73). 511

Statistical analyses. Depending on normal or not normal distribution of data, t-test or Mann-512

Whitney test were used for sex comparison, ANOVA or Kruskal-Wallis for category comparison (H 513

vs. CY vs. FD) (Table S11). Tukey or Dunn post-hoc tests were used following ANOVA or 514

Kruskal-Wallis, respectively. Whenever no sex-related differences were recorded within the same 515

category, female and male data were pooled. SIGMAPLOT 11 (Systat Software) was used. 516

517

518 on April 11, 2018 by INRA - Institut National de la Recherche Agronomique

http://iai.asm.org/

(23)

22 Contributions of authors

519

Design of experiments: LG SA CM MR MV DB; RNAseq, bioinformatic analysis and differential 520

gene expression profiling: SA; RT-qPCR validation, statistical analysis and protein expression: LG; 521

phenoloxidase enzymatic activity: LG CM MPes; pigmentation assay: MR MV SA; 522

immunocompetence assay: CM NAB MPD; movement assay: SA MR MV MPes LG; respiration 523

assay: LG WC; insect rearing and plant production: MPeg. LG SA MR MV CM wrote the paper 524

and all authors reviewed the manuscript. 525

526

Conflict of interest 527

The authors declare that they have no conflicts of interest with the contents of this article. 528

529

ACKNOWLEDGMENTS 530

The Authors thank Brigitte Bataillerfor helping with immunocompetence assay, Francesco 531

Pennacchio and Gennaro Di Prisco, University of Naples Federico II, for helpful discussion and 532

suggestions. 533

This work was part of the ‘FitoDigIt’ Project funded by Fondazione Cassa di Risparmio di Torino, 534

Torino (Italy), within the ‘Richieste Ordinarie 2014’ and ‘Richieste Ordinarie 2015’ calls. MR and 535

MPeg were supported by a fellowship funded by the following grant-making foundations: 536

Fondazione Cassa di Risparmio di Cuneo, Fondazione Cassa di Risparmio di Torino, and 537

Fondazione Cassa di Risparmio di Asti in the framework of the INTEFLAVI project. The funders 538

had no role in study design, data collection and interpretation, or the decision to submit the work for 539 publication. 540 541 REFERENCES 542

1. Marcone C. 2014. Molecular biology and pathogenicity of phytoplasmas. Ann Appl Biol 165:199–221. 543

http://iai.asm.org/

(24)

23

2. Maejima K, Oshima K, Namba S. 2014. Exploring the phytoplasmas, plant pathogenic bacteria. J Gen 544

Plant Pathol 80:210–221. 545

3. Oshima K, Maejima K, Namba S. 2013. Genomic and evolutionary aspects of phytoplasmas. Front 546

Microbiol 4:230. 547

4. Pacifico D, Galetto L, Rashidi M, Abbà S, Palmano S, Firrao G, Bosco D, Marzachì C. 2015. Decreasing 548

global transcript levels over time suggest that phytoplasma cells enter stationary phase during plant 549

and insect colonization. Appl Environ Microbiol 81:2591–2602. 550

5. Oshima K, Ishii Y, Kakizawa S, Sugawara K, Neriya Y, Himeno M, Minato N, Miura C, Shiraishi T, Yamaji 551

Y, Namba S. 2011. Dramatic transcriptional changes in an intracellular parasite enable host switching 552

between plant and insect. PLoS ONE 6:e23242. 553

6. Caudwell A, Kuszala C, Larrue J, Bachelier J. 1972. Transmission de la Flavescence dorée de la fève à la 554

fève par des cicadelles des genres Euscelis et Euscelidius. Ann Phytopathol No. hors série:181–189. 555

7. Rashidi M, D’Amelio R, Galetto L, Marzachì C, Bosco D. 2014. Interactive transmission of two 556

phytoplasmas by the vector insect. Ann Appl Biol 165:404–413. 557

8. Bressan A, Clair D, Sémétey O, Boudon-Padieu É. 2005. Effect of two strains of Flavescence dorée 558

phytoplasma on the survival and fecundity of the experimental leafhopper vector Euscelidius 559

variegatus Kirschbaum. J Invertebr Pathol 89:144–149. 560

9. Bosco D, Marzachì C. 2016. Insect transmission of phytoplasmas, p. 319–327. In Brown, JK (ed.), 561

Vector-mediated transmission of plant pathogens. APS Press, St. Paul, Minnesota. 562

10. Abbà S, Galetto L, Carle P, Carrère S, Delledonne M, Foissac X, Palmano S, Veratti F, Marzachì C. 2014. 563

RNA-Seq profile of flavescence dorée phytoplasma in grapevine. BMC Genomics 15:1088. 564

http://iai.asm.org/

(25)

24

11. Galetto L, Bosco D, Balestrini R, Genre A, Fletcher J, Marzachì C. 2011. The major antigenic membrane 565

protein of ‘Candidatus Phytoplasma asteris’ selectively interacts with ATP synthase and actin of 566

leafhopper vectors. PLoS ONE 6:e22571. 567

12. Rashidi M, Galetto L, Bosco D, Bulgarelli A, Vallino M, Veratti F, Marzachì C. 2015. Role of the major 568

antigenic membrane protein in phytoplasma transmission by two insect vector species. BMC 569

Microbiol 15:193. 570

13. Arricau-Bouvery N, Duret S, Dubrana M-P, Batailler B, Desqué D, Béven L, Danet J-L, Monticone M, 571

Bosco D, Malembic-Maher S, Foissac X. 2018. Variable membrane protein A of flavescence dorée 572

phytoplasma binds the midgut perimicrovillar membrane of Euscelidius variegatus and promotes 573

adhesion to its epithelial cells. Appl Environ Microbiol AEM.02487-17. 574

14. Howell L, Sampson CJ, Xavier MJ, Bolukbasi E, Heck MMS, Williams MJ. 2012. A directed miniscreen 575

for genes involved in the Drosophila anti-parasitoid immune response. Immunogenetics 64:155–161. 576

15. Armitage SAO, Siva-Jothy MT. 2005. Immune function responds to selection for cuticular colour in 577

Tenebrio molitor. Heredity 94:650–656. 578

16. Morano I. 2013. Muscles and Motility, p. 461–478. In Galizia, CG, Lledo, P-M (eds.), Neurosciences - 579

From Molecule to Behavior: a university textbook. Springer Berlin Heidelberg, Berlin, Heidelberg. 580

17. Waniek PJ, Pacheco Costa JE, Jansen AM, Costa J, Araújo CAC. 2012. Cathepsin L of Triatoma 581

brasiliensis (Reduviidae, Triatominae): sequence characterization, expression pattern and 582

zymography. J Insect Physiol 58:178–187. 583

18. Renoz F, Noël C, Errachid A, Foray V, Hance T. 2015. Infection dynamic of symbiotic bacteria in the 584

pea aphid Acyrthosiphon pisum gut and host immune response at the early steps in the infection 585

process. PLOS ONE 10:e0122099. 586

19. Rolff J. 2002. Bateman’s principle and immunity. Proc R Soc B Biol Sci 269:867–872. 587

http://iai.asm.org/

(26)

25

20. Sheridan LAD, Poulin R, Ward DF, Zuk M. 2000. Sex differences in parasitic infections among 588

arthropod hosts: is there a male bias? Oikos 88:327–334. 589

21. Blaauw BR, Jones VP, Nielsen AL. 2016. Utilizing immunomarking techniques to track Halyomorpha 590

halys (Hemiptera: Pentatomidae) movement and distribution within a peach orchard. PeerJ 4:e1997. 591

22. Matsumoto Y, Suetsugu Y, Nakamura M, Hattori M. 2014. Transcriptome analysis of the salivary 592

glands of Nephotettix cincticeps (Uhler). J Insect Physiol 71:170–176. 593

23. DeLay B, Mamidala P, Wijeratne A, Wijeratne S, Mittapalli O, Wang J, Lamp W. 2012. Transcriptome 594

analysis of the salivary glands of potato leafhopper, Empoasca fabae. J Insect Physiol 58:1626–1634. 595

24. Shao E, Lin G, Liu S, Ma X, Chen M, Lin L, Wu S, Sha L, Liu Z, Hu X, Guan X, Zhang L. 2017. Identification 596

of transcripts involved in digestion, detoxification and immune response from transcriptome of 597

Empoasca vitis (Hemiptera: Cicadellidae) nymphs. Genomics 109:58–66. 598

25. Chen Y, Cassone BJ, Bai X, Redinbaugh MG, Michel AP. 2012. Transcriptome of the plant virus vector 599

Graminella nigrifrons, and the molecular interactions of Maize fine streak rhabdovirus transmission. 600

PLoS ONE 7:e40613. 601

26. Nandety RS, Kamita SG, Hammock BD, Falk BW. 2013. Sequencing and de novo assembly of the 602

transcriptome of the glassy-winged sharpshooter (Homalodisca vitripennis). PLoS ONE 8:e81681. 603

27. Asgharian H, Chang PL, Mazzoglio PJ, Negri I. 2014. Wolbachia is not all about sex: male-feminizing 604

Wolbachia alters the leafhopper Zyginidia pullula transcriptome in a mainly sex-independent manner. 605

Front Microbiol 5:430. 606

28. Cassone BJ, Cisneros Carter FM, Michel AP, Stewart LR, Redinbaugh MG. 2014. Genetic insights into 607

Graminella nigrifrons competence for Maize fine streak virus infection and transmission. PLoS ONE 608

9:e113529. 609

http://iai.asm.org/

(27)

26

29. Kumaresan V, Harikrishnan R, Arockiaraj J. 2015. A potential Kazal-type serine protease inhibitor 610

involves in kinetics of protease inhibition and bacteriostatic activity. Fish Shellfish Immunol 42:430– 611

438. 612

30. Kim BY, Lee KS, Zou FM, Wan H, Choi YS, Yoon HJ, Kwon HW, Je YH, Jin BR. 2013. Antimicrobial activity 613

of a honeybee (Apis cerana) venom Kazal-type serine protease inhibitor. Toxicon 76:110–117. 614

31. Monwan W, Amparyup P, Tassanakajon A. 2017. A snake-like serine proteinase ( Pm Snake) activates 615

prophenoloxidase-activating system in black tiger shrimp Penaeus monodon. Dev Comp Immunol 616

67:229–238. 617

32. Barillas-Mury C. 2007. CLIP proteases and Plasmodium melanization in Anopheles gambiae. Trends 618

Parasitol 23:297–299. 619

33. Liu Y, Hou F, He S, Qian Z, Wang X, Mao A, Sun C, Liu X. 2014. Identification, characterization and 620

functional analysis of a serine protease inhibitor (Lvserpin) from the Pacific white shrimp, Litopenaeus 621

vannamei. Dev Comp Immunol 43:35–46. 622

34. Dubovskiy I, Kryukova N, Glupov V, Ratcliffe N. 2016. Encapsulation and nodulation in insects. 623

Invertebr Surviv J 13:229–246. 624

35. Cornet S, Biard C, Moret Y. 2009. Variation in immune defence among populations of Gammarus 625

pulex (Crustacea: Amphipoda). Oecologia 159:257–269. 626

36. Eliautout R, Dubrana M-P, Vincent-Monégat C, Vallier A, Braquart-Varnier C, Poirié M, Saillard C, 627

Heddi A, Arricau-Bouvery N. 2016. Immune response and survival of Circulifer haematoceps to 628

Spiroplasma citri infection requires expression of the gene hexamerin. Dev Comp Immunol 54:7–19. 629

37. Huang P-H, Lu S-C, Yang S-H, Cai P-S, Lo C-F, Chang L-K. 2014. Regulation of the immediate-early genes 630

of white spot syndrome virus by Litopenaeus vannamei kruppel-like factor (LvKLF). Dev Comp 631

Immunol 46:364–372. 632

http://iai.asm.org/

(28)

27

38. Castagnola A, Jurat-Fuentes JL. 2016. Intestinal regeneration as an insect resistance mechanism to 633

entomopathogenic bacteria. Curr Opin Insect Sci 15:104–110. 634

39. Chen X, Yao P, Chu X, Hao L, Guo X, Xu B. 2015. Isolation of arginine kinase from Apis cerana cerana 635

and its possible involvement in response to adverse stress. Cell Stress Chaperones 20:169–183. 636

40. Jiang S, Jia Z, Chen H, Wang L, Song L. 2016. The modulation of haemolymph arginine kinase on the 637

extracellular ATP induced bactericidal immune responses in the Pacific oyster Crassostrea gigas. Fish 638

Shellfish Immunol 54:282–293. 639

41. Ma C, Gao Q, Liang Y, Li C, Liu C, Huang J. 2016. Shrimp arginine kinase being a binding protein of 640

WSSV envelope protein VP31. Chin J Oceanol Limnol 34:1287–1296. 641

42. Sun J. 2012. Roles of cellular redox factors in pathogen and toxin entry in the endocytic pathways, p. 642

61–90. In Ceresa, B (ed.), Molecular regulation of endocytosis. InTech. 643

43. Kongton K, McCall K, Phongdara A. 2014. Identification of gamma-interferon-inducible lysosomal thiol 644

reductase (GILT) homologues in the fruit fly Drosophila melanogaster. Dev Comp Immunol 44:389– 645

396. 646

44. Ren C, Chen T, Jiang X, Luo X, Wang Y, Hu C. 2015. The first echinoderm gamma-interferon-inducible 647

lysosomal thiol reductase (GILT) identified from sea cucumber (Stichopus monotuberculatus). Fish 648

Shellfish Immunol 42:41–49. 649

45. Al-Quadan T, Price CT, London N, Schueler-Furman O, AbuKwaik Y. 2011. Anchoring of bacterial 650

effectors to host membranes through host-mediated lipidation by prenylation: a common paradigm. 651

Trends Microbiol 19:573–579. 652

46. Hicks SW, Galán JE. 2013. Exploitation of eukaryotic subcellular targeting mechanisms by bacterial 653

effectors. Nat Rev Microbiol 11:316–326. 654

http://iai.asm.org/

(29)

28

47. Anabestani A, Izadpanah K, Abbà S, Galetto L, Ghorbani A, Palmano S, Siampour M, Veratti F, 655

Marzachì C. 2017. Identification of putative effector genes and their transcripts in three strains 656

related to ‘Candidatus Phytoplasma aurantifolia’. Microbiol Res 199:57–66. 657

48. Mayans O, Benian GM, Simkovic F, Rigden DJ. 2013. Mechanistic and functional diversity in the 658

mechanosensory kinases of the titin-like family. Biochem Soc Trans 41:1066–1071. 659

49. Velthuis AJW, Bagowski CP. 2007. PDZ and LIM domain-encoding genes: molecular interactions and 660

their role in development. Sci World J 7:1470–1492. 661

50. Rossi D, Sorrentino V. 2002. Molecular genetics of ryanodine receptors Ca2+_{-release channels. Cell}

662

Calcium 32:307–319. 663

51. Takekura H, Franzini-Armstrong C. 2002. The structure of Ca2+ release units in arthropod body muscle 664

indicates an indirect mechanism for excitation-contraction coupling. Biophys J 83:2742–2753. 665

52. Ji T, Yin L, Liu Z, Shen F, Shen J. 2014. High-throughput sequencing identification of genes involved 666

with Varroa destructor resistance in the eastern honeybee, Apis cerana. Genet Mol Res 13:9086– 667

9096. 668

53. Ayme-Southgate A, Feldman S, Fulmer D. 2015. Myofilament proteins in the synchronous flight 669

muscles of Manduca sexta show both similarities and differences to Drosophila melanogaster. Insect 670

Biochem Mol Biol 62:174–182. 671

54. Martini X, Hoffmann M, Coy MR, Stelinski LL, Pelz-Stelinski KS. 2015. Infection of an insect vector with 672

a bacterial plant pathogen increases its propensity for dispersal. PLOS ONE 10:e0129373. 673

55. Wang G, Zhang J, Shen Y, Zheng Q, Feng M, Xiang X, Wu X. 2015. Transcriptome analysis of the brain 674

of the silkworm Bombyx mori infected with Bombyx mori nucleopolyhedrovirus: a new insight into the 675

molecular mechanism of enhanced locomotor activity induced by viral infection. J Invertebr Pathol 676

128:37–43. 677

http://iai.asm.org/

(30)

29

56. Saikhedkar N, Summanwar A, Joshi R, Giri A. 2015. Cathepsins of lepidopteran insects: aspects and 678

prospects. Insect Biochem Mol Biol 64:51–59. 679

57. Katunuma N. 2010. Posttranslational processing and modification of cathepsins and cystatins. J Signal 680

Transduct 2010:1–8. 681

58. Futahashi R, Tanaka K, Tanahashi M, Nikoh N, Kikuchi Y, Lee BL, Fukatsu T. 2013. Gene expression in 682

gut symbiotic organ of stinkbug affected by extracellular bacterial symbiont. PLoS ONE 8:e64557. 683

59. Abbà S, Galetto L, Vallino M, Rossi M, Turina M, Sicard A, Marzachì C. 2017. Genome sequence, 684

prevalence and quantification of the first iflavirus identified in a phytoplasma insect vector. Arch Virol 685

162:799–809. 686

60. Bolger AM, Lohse M, Usadel B. 2014. Trimmomatic: a flexible trimmer for Illumina sequence data. 687

Bioinformatics 30:2114–2120. 688

61. FastQC. 2016. https://www.bioinformatics.babraham.ac.uk/projects/fastqc/. 689

62. Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden J, Couger MB, Eccles D, Li B, 690

Lieber M, MacManes MD, Ott M, Orvis J, Pochet N, Strozzi F, Weeks N, Westerman R, William T, 691

Dewey CN, Henschel R, LeDuc RD, Friedman N, Regev A. 2013. De novo transcript sequence 692

reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nat 693

Protoc 8:1494–1512. 694

63. Li W, Godzik A. 2006. Cd-hit: a fast program for clustering and comparing large sets of protein or 695

nucleotide sequences. Bioinformatics 22:1658–1659. 696

64. Powell S, Szklarczyk D, Trachana K, Roth A, Kuhn M, Muller J, Arnold R, Rattei T, Letunic I, Doerks T, 697

Jensen LJ, von Mering C, Bork P. 2012. eggNOG v3.0: orthologous groups covering 1133 organisms at 698

41 different taxonomic ranges. Nucleic Acids Res 40:D284–D289. 699

65. Transdecoder. 2016. https://transdecoder.github.io/. 700

http://iai.asm.org/